The corneal endothelium plays a pivotal role in maintaining corneal transparency. Unlike in other species, the human corneal endothelium is notorious for its limited proliferative capacity in vivo after diseases, injury, aging, and surgery Persistent corneal endothelial dysfunction leads to sight-threatening bullous keratopathy. Presently, the only solution to restore vision in eyes inflicted with bullous keratopathy relies upon transplantation of a cadaver donor cornea containing a healthy corneal endothelium. Due to a severe global shortage of donor corneas, in conjunction with an increasing trend toward transplanting only the corneal endothelium in procedures collectively termed endothelial keratoplasties, it is timely and paramount to develop a tissue engineering strategy to produce surgical grafts containing human corneal endothelial cells (HCECs). Using our reported in vitro model system, in which the mitotic block is mediated by contact inhibition when cell junctions mature, we have shown that the conventional engineering methods using EDTA/bFGF to generate single HCECs activates -catenin/Wnt signaling and the loss of the normal HCEC phenotype to endothelial-mesenchymal transition (EMT). In contrast, our novel engineering method based on transient knockdown by p120 catenin (p120) and Kaiso siRNAs unlocks the mitotic block by activating p120/Kaiso signaling but not -catenin/Wnt signaling. We have further optimized this p120-Kaiso knockdown regimen by switching to a serum-free medium containing bFGF and LIF and discovered that our method further activates RohA-ROCK-canonical BMP signaling to reprogram HCECs to neural-crest like progenitors which proliferates to maintain the normal HCEC phenotype without EMT. Consequently, our novel tissue engineering technology can successfully produce from Descemet membrane stripped from 1/8 of the corneoscleral rim (normally discarded after conventional corneal transplantation) one HCEC monolayer with a hexagonal shape, comparable in vivo cell density, and an average size of 11.0 0.6 mm in diameter. That is, the technology will add at least an additional 8 transplantable grafts per one donor cornea. In this Phase II application, we propose to establish reproducible GMP engineering of HCEC grafts on an implantable collagen membrane, with a new packing system to transport these grafts (Aim 1), and to examine the safety and efficacy of these engineered HCEC grafts through the surgical procedure of DMEK in an in vivo NIH mini pig model of endothelial dysfunction (Aim 2). Completion of these two Aims will allow the Company to gather sufficient pre-clinical data needed for an IND submission to the FDA. Ultimately, the Company can capture a unique market opportunity by fulfilling an unmet global need. One day, this new tissue engineering technology can also be deployed to engineer other similar monolayer tissues such as retinal pigment epithelium (RPE) for submacular transplantation in treating retinal blinding diseases characterized by dysfunctional RPE. Furthermore, successful commercialization of this technology will stimulate the scientific community to re-think how contact inhibition can safely be perturbed to our benefit, i.e., by maintaining the normal phenotype, and whether this new regenerative approach can circumvent the need to reprogramming directly from embryonic stem cells or induced pluripotent stem cells.